Emerging precision flight applications are generating control surface actuator performance requirements that exceed the capability of electrical and hydraulic actuator technology. For example, a vehicle performing a carrier landing must keep within ±12 in of the projected landing approach flight trajectory near touch-down in order to avoid a wave-off and to successfully hook the arrestor cable. The control requirement for naval style mid-air refueling is even more demanding than the requirement for the carrier landing. In the refueling scenario, a low-on-fuel vehicle has to intercept a fuel hose basket to within ±6 inches while the tanker maintains a constant flight trajectory. While not quite as demanding on the aircraft as the naval scenario, Air Force style refueling still requires the low-on-fuel vehicle to maintain its position relative to the tanker within ±4 ft. Meanwhile, the boom operator aboard the tanker has to “fly” the 40-ft long boom to within ±6 inches of the bobbing fuel probe on the low-on-fuel aircraft. Thus, the ruddevators (i.e. small wings on the refueling boom) must be controlled rather accurately. The need for an aircraft to hook a cable, or intercept a refuel drogue, or the like translates into control surface deflections so small that they lie beneath the range of the minimum achievable backlash in available actuator gearheads and linkages. Moreover, static friction in the control systems causes resistance to the small movements associated with the vernier attitude control required for such demanding operations. These nonlinearities limit the size of the smallest control surface movement that can be reliably repeated. Even perfect actuator position feedback along with unlimited control system throughput cannot avoid the detrimental effects of the nonlinearities.
Automatic control, during operations requiring vernier control, is therefore not currently achievable. Manned vehicles have succeeded in carrying out these difficult piloting tasks only because pilots adapt to nonlinearity (e.g. the friction and backlash in the control system) in subtle ways that heretofore have defied duplication in software. The nonlinearities impeding the automation of these tasks, though, cannot be eliminated because they are inherent in the available actuator technology.
Attempting to work around the resulting dead band using smaller rate command steps is possible, at least theoretically. Doing so would require boosting actuator loop gain and controlling pitch rate; roll rate, or yaw rate primarily through pulse width modulation of the actuator commands. However, this solution invites limit-cycle oscillations (e.g. actuator “hunting”) that reduce actuator life through excessive fatigue. Other previous attempts at controlling pitch, roll, and yaw rates include morphing airfoils into new shapes by using electrical or hydraulic actuators that drive hinged flaps or variable camber surface-bending mechanisms. Although these previous control systems have performed “coarse” attitude control satisfactorily, in the future, much faster, higher resolution control will be needed than can be achieved with this technology.
Another work-around might involve using an electro-hydraulic “coarse” position actuator supplemented with a smaller “fine” position control actuator that has higher bandwidth capability. The smaller actuator's loop gain could also be increased so that the vehicle will follow the fine actuator the majority of the time. The coarse actuator would therefore quietly serve in a secondary role for causing relatively large attitude changes. Unfortunately, all of the problems previously discussed with respect to the larger actuator would also apply to the fine actuator. Plus, because of the high gain required for the small actuator, instabilities could develop in the “fine” control system,
Thus, a need exists for a mobile platform attitude control system that provides vernier attitude control sufficient to maintain a mobile platform within about six inches, or less, of a target trajectory.